ICAAC 2003 coverage

Drug resistance guide being pondered by CDC

As HIV drug resistance rises, surveillance needed

Guidelines released this year from the U.S. Department of Health and Human Services (DHHS)1 and the International AIDS Society-USA2 (IAS) recommend resistance testing for all recently infected patients beginning treatment.

However, for the first time, both groups also suggest considering resistance testing for drug-naïve patients whose infections are not recent, particularly if the prevalence of resistance in the patient’s area is 5% or greater.

While that would appear to be a sound recommendation, the question many clinicians might ask is, "How will we know what the resistance prevalence is in our area?"

The Centers for Disease Control and Prevention (CDC) is collaborating with health departments to begin resistance surveillance programs in 18 cities and states, partly for the purpose of answering that question.

"At this point, not many areas have that surveillance information to give to clinicians," says Diane Bennett, MD, a medical epidemiologist and the coordinator of Antiretroviral Drug Resistance Surveillance at the CDC.

"For years, we have had resistance surveillance for many other infectious diseases, but not HIV," she says. "But resistant HIV strains are being transmitted in the U.S. to people who have never taken antiretroviral drugs themselves."

Previously, the CDC collaborated with health departments in a 10-city antiretroviral resistance surveillance study in drug-naïve people newly diagnosed with HIV from 1997 to 2001, Bennett adds.

Data show that 11.5% of those recently infected with HIV, and 7.4% of those whose infections were not recent had HIV resistance to one or more antiretroviral drugs, she says.

In recently infected people, resistance rose from 7.7% in 1997-1998 to 12.7% in 2000-2001. Among people whose infections were not recent, resistance rose from 4.4% to 8.2%, Bennett adds.

"But even though 1,082 people were tested, this study, like all studies so far, wasn’t big enough to give precise estimates of resistance prevalence," she says.

Over the past year, health departments in four areas have collaborated with the CDC to pilot antiretroviral drug resistance surveillance. The project is being expanded to 14 new areas for a total of 18, Bennett states.

In the new surveillance system, resistance testing will be free for anyone newly diagnosed with HIV. Each HIV strain will be fully genotyped for resistance testing with standard sequencing on 333 positions in reverse transcriptase and 100 in protease sites, she explains. "But for a real picture of what’s going on in the U.S., we should add even more areas, and that could be very expensive."

Since genotyping tests cost roughly $290 to $400 each, and since there are an estimated 40,000 new HIV infections a year, full genotyping is an expensive method to conduct antiretroviral resistance surveillance when looking at the United States as a whole, she says. "But right now, it’s the cheapest method we have."

So while the surveillance study is under way, the CDC also is assessing a potential screening algorithm for drug resistance surveillance, Bennett continues.

Complex statistical methods were used to come up with 10 "indicator" mutations that could be used for screening in a surveillance system.

In a study presented at the 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy, held recently in Chicago, the CDC showed that its screening algorithm identified 96% of the antiretroviral resistant strains in the 1997-2001 U.S. data set, but that the algorithm will need to be validated against other data sets.3

"What we’re going to be doing is validating the algorithm against other data sets that have been fully genotyped to see if we had used the algorithm in those past data sets how many [drug resistant samples] we would have missed," Bennett explains.

The screening algorithm clearly wouldn’t be accurate enough to be used for clinical purposes in the treatment of individual patients, but it may be accurate enough for surveillance of a large population, she notes.

The CDC’s plan, if the algorithm proves to be useful for surveillance, is to develop point mutation assays to screen specific indicator positions in the reverse transcriptase and protease positions.3

The laboratory assays also would require full testing and validation before the strategy could be considered as a surveillance tool.

"If both the algorithm and the assays work, it would be possible to screen a very large number of HIV strains from newly diagnosed people, first using the point mutation assays," Bennett says. "Any strain that had one of the indicator positions would go on for full genotyping, and this strategy would be cheaper than fully genotyping every strain."

However, the algorithm would have to be continually updated, and this is only one of several potential surveillance strategies, she adds. Provid-ing antiretroviral resistance surveillance data to various U.S. regions would provide several public health benefits, including providing valuable information to support the development of updated DHHS and IAS guidelines, Bennett says.

It also would give clinicians crucial information when they are considering post-exposure prophylaxis in HIV exposure cases where there isn’t time to wait for a genotyping test, she says.

"Surveillance of HIV resistance needs to be done to support public health guidelines and for a variety of clinical and public health reasons," Bennett says. "Whether or not the algorithm proves useful, some form of expanded resistance surveillance is needed."


1. Guidelines for the Use of Antiretroviral Drugs in HIV-1 Infected Adults and Adolescents, July 14, 2003. Available at: http://www.aidsinfo.nih.gov/guidelines/.

2. Antiretroviral drug resistance testing in adults infected with HIV Type 1:2003. Recommendations of an International AIDS Society — USA Panel, Clin Infect Dis 2003; 37:113-128.

3. Bennett DE, Byers R, Johnson J, et al. A screening algorithm for surveillance of antiretroviral drug resistance among individuals newly diagnosed with HIV. Poster presented at the 43rd Interscience Conference on Antimicrobial Agents and Chemotherapy. Chicago; Sept. 14-17, 2003. H-909.